This month, we welcomed two of the new Moorfields trainees and Esin, a paediatric pathologist from The Royal London Hospital. We covered corneal and retinal histology and chatted about how some tumours may have deceptively bland cytology. Cases included a metastatic carcinoma with unknown primary, a retinal detachment and a cavernous haemangioma. We had a discussion about non- or pre-surgical treatments of haemangiomas and I had to admit I don’t know whether orbital cavernous haemangiomas can be pre-treated. Vascular malformations can be embolised preoperatively to shrink them (see my post from October 2016 for examples) and there are studies on topical and oral beta-blockers for infantile haemangiomas. I found articles here, here and here and a GOSH guidance sheet.
For the histopathologists, and ophthalmologists with an interest in ocular oncology, please note that the Eye Pathology Blue Book will be available soon, probably sometime in October. Non-histopathologists may not have come across the WHO Classification of Tumours series, also known as the Blue Books. The series is a set of reference books which provide classification and diagnostic criteria for benign and malignant tumours. You can read more in this article by Ian Cree, who was one of my predecessors at the UCL Institute of Ophthalmology.
Here are some of the cases we discussed this month.
Case 1 – Enucleation specimen for intraocular mass
This low power view shows the sclera with an intraocular tumour (lower part of the image). The feature of note is the structure in the outer part of the sclera.
Medium power shows a nerve bundle.
I suspect that the section has sampled part of an intrascleral nerve loop (of Axenfeld). This is sometimes noted clinically as a brown spot on the sclera. It can be a pitfall as it can be misdiagnosed as extraocular spread of intraocular melanoma. I found a couple of clinical images here and here.
High power view of the tumour shows epithelioid cells (centrally) and a mitosis (lower right). This is malignant melanoma within the ciliary body and choroid.
Case 2 – Exenteration specimen for orbital tumour
This low power view shows an infiltrating tumour with nests of basaloid cells. There are two patterns here which I will expand on below.
This field shows rather tubular architecture – as on the left of the previous figure.
In contrast, this field shows the classic cribriform pattern of adenoid cystic carcinoma.
Adenoid cystic carcinoma can have a number of architectural patterns (cribriform, tubular and solid). The cribriform pattern is the commonest, and most easily recognisable. If you’re interested in epithelial lacrimal gland tumours, here’s a nice free access review article by von Holstein.
Case 3 – Orbital tumour
Here’s another adenoid cystic carcinoma.
Again, on low power, the cribriform architecture is striking.
Higher power shows the pseudocystic spaces, some of which are incomplete. They contain myxoid or hyaline material.
A whole slide image (WSI) from this case will be made available to view online as part of the Blue Book resources.
Case 4 – Back to more prosaic matters
This is a cornea.
This low power view of full thickness cornea shows epithelial bullous lifting (from the underlying Bowman’s layer) as well as surface blebbing.
On higher power, we can better appreciate the surface blebbing (*) and tongues of regenerated epithelium basally (x).
PAS stain of the same field shows corresponding basement membrane disorganisation.
This PAS stain shows Descemet’s membrane to be normal–no guttae and no thickening–but there is a sheet of fibrous tissue posteriorly, obliterating the endothelium.
So this is bullous keratopathy and retrocorneal fibrosis.
That’s all for this month. Next month’s session will be on Wednesday 17 October. See you then!
2 thoughts on “Microscopy September 2018”
Thanks for these very good cases
Catherine. Did the adenoid cystic cases show perineural invasion?
One of them did, but I didn’t manage to find PNI in the other. Even though it’s a classic feature (and a very useful clue), I’ve found it quite difficult to detect. On occasions where I wasn’t completely confident of the diagnosis, I’ve sometimes looked at immunohistochemistry for S100, just to make PNI easier to find.